This project, a response to notice number NOT-OD-10-032 and notice title NIH Announces the Availability of Recovery Act Funds for Competitive Revision Applications (R01, R03, R15, R21/R33, and R37) through the NIH Basic Behavioral and Social Science Opportunity Network (OppNet), focuses on approaches to data analysis that allow researchers to investigate underlying mechanisms and processes of mental health and behavioral problems that are targeted by prevention and treatment programs in a wide range of behavioral research areas. Specifically, the work will evaluate statistical modeling approaches that can be used to examine processes targeted by prevention and treatment programs (mediators) when outcomes are measured as time until the onset of an event (survival data), and thereby provide guidance to researchers who consider using these approaches. The parent grant focuses on treatment for major depressive disorder, but as part of the proposal, collaboration will be established with a leading methodological and health promotion and disease prevention researcher. Furthermore, these approaches would apply to mental health disorders other than depression, to substance abuse, and to health behaviors (e.g., weight control), and need not be utilized in the context of applied research, though they will be in the case of the parent grant. The parent grant, Prevention of Recurrence in Depression with Drugs and Cognitive Therapy, is an NIMH-funded multi-site trial in which depressed outpatients have been randomized to antidepressant medications (ADM) alone or to ADM and cognitive therapy (CT). It was designed with power sufficient to: 1) detect benefits that might accompany the addition of CT to ADM with respect to treatment response and the prevention of recurrence, and 2) explore predictors and underlying processes that are targeted by the treatments, and assess their relationships to symptom improvement (a basic behavioral and social sciences research goal), potentially elucidating how additional benefits might accrue from adding CT and providing information about how to improve treatments for depression. Measures of multiple kinds of cognitive changes and treatment processes are being collected in order to test hypotheses about these constructs as underlying mechanisms of symptom improvement. The proposed simulation work, which is not included in the original parent grant, will help to ensure that the best approaches are used to achieve this aim. PUBLIC HEALTH RELEVANCE: Depression is one of the most prevalent and debilitating of the psychiatric disorders. By yielding estimates of the benefit of adding psychotherapy to medication therapy in the treatment of depression, and identifying subgroups of patients who most benefit from combined treatment, findings from this project will be able to guide clinicians in selecting treatments, and improving outcomes. In addition, mechanisms and processes that influence outcome will be studied, which will have the potential to lead to new treatment approaches and further improved outcomes.